PRICING
AI security pricing that scales with your actual agent risk
Discover your AI attack surface for free. Pay only for the agents, workflows, MCP servers, and LLM applications Intertrace actively protects with runtime enforcement, evidence, and Guardian monitoring.
Most teams start with 10 to 30 protected agents, validate risk reduction in production, then expand across providers, clouds, tools, and business units.
Pay for protected risk, not passive inventory
Intertrace does not charge you just to discover AI assets. Discovery helps you map the attack surface first. Pricing begins when an agent, workflow, MCP server, or LLM application is actively monitored, governed, or enforced by Intertrace.
That means you can discover everything for free, protect only what matters, and scale when your AI footprint and risk profile grow.
Discover for free
Map agents, models, MCP servers, and high-risk workflows before you commit to protection spend.
Protect what matters
Pay only for systems you actively enforce—runtime policy, evidence, Guardian monitoring, and governance workloads.
Expand with risk
Add agents, volume, integrations, and deployment controls as AI adoption spreads across teams and clouds.
The cost of undefended AI agents
One compromised AI agent can expose customer data, misuse internal tools, leak secrets, or create audit evidence gaps. Intertrace helps security teams prioritize and protect the agents that create real enterprise risk—so pricing is anchored to outcomes, not seat count.
Why Intertrace is priced by protected agent
AI risk is not created equally. A low-risk internal summarizer should not cost the same as a tool-using agent with customer data, payment access, browser automation, or broad MCP permissions. Intertrace prices around the systems that actually create enterprise risk.
- Prevent prompt injection and tool abuse before production impact.
- Reduce manual AI security review time with continuous runtime signals.
- Generate evidence for security, compliance, and audit teams.
- Prioritize agents that can access sensitive data or take external action.
- Avoid paying for passive inventory—you fund active protection.
What counts as a protected agent?
A protected agent is any distinct AI application, autonomous agent, RAG workflow, MCP connected agent, LLM powered workflow, or production AI service that Intertrace monitors, governs, or enforces in production or controlled environments.
| Example | Status | Reason |
|---|---|---|
| Customer Support Agent | Counts as protected agent | Distinct agent with customer data access and tool use. |
| Sales RAG Assistant | Counts as protected agent | Distinct RAG workflow retrieving sensitive sales content. |
| MCP Browser Agent | Counts as protected agent | MCP connected agent with external browser capability. |
| Finance Assistant | Counts as protected agent | Agent with billing or financial data access. |
| Internal API LLM workflow | Counts as protected agent | LLM powered workflow exposed through an internal or production API. |
| Unprotected discovered AI asset | Does not count until protected | Inventory discovery is free until Intertrace actively protects, governs, or enforces it. |
If an AI system has its own identity, purpose, tool set, data access pattern, runtime policy, or production deployment, it typically counts as a protected agent.
Annual protection tiers
Start with Production Validation or a free snapshot if you need proof before a multi-year agreement. Most multi-team buyers anchor on Scale once MCP, SIEM, and multi-provider coverage matter.
Annual contracts only. Paid POCs are available for teams that need validation before committing.
Paid production POC
Production Validation
$25K to $50K
Credited toward annual contract when you convert.
For teams that need to validate Intertrace before an annual purchase—without diluting production-tier positioning.
- Up to 3 protected agents
- 1M prompts
- One provider
- Basic findings
- Evidence export
- Guardian Lite
- End-of-POC report
Free assessment
Free AI Attack Surface Snapshot
Discover exposed AI agents, tools, MCP servers, and high-risk workflows before you buy. Snapshot assessments are scoped with our team—no self-serve free tier that undermines enterprise positioning.
Request free snapshotLaunch
Starting at $60K per year
Prove production AI protection on your first high-risk agents.
Best for
- First production deployment
- One provider
- One cloud
- Security validation
- SOC 2 evidence needs
Not ideal for
- Multi-provider environments
- MCP-heavy agents
- SIEM-integrated security operations
- Cross-team AI governance
Outcomes
- Prove production AI protection on a focused set of high-risk agents.
- Detect prompt injection and jailbreak attempts in real time with enforceable policy paths.
- Find PII and secrets in agent traffic and export evidence for review.
- Map blast radius and behavioral signals to prioritize remediations.
- Generate audit- and SOC 2–oriented evidence packs for GRC reviewers.
- Validate in production with a 30-day POC on annual agreements.
MCP Gateway, Tool Guard, multi-provider coverage, and advanced compliance paths are available as expansion options or in Scale.
Scale
Most common enterprise starting point
Starting at $180K per year
Typically selected by teams with 10 to 30 production AI agents.
The default plan for teams securing AI agents across multiple providers, tools, MCP servers, and business workflows.
Outcomes
- Protect tool-using agents across multiple teams and business units.
- Enforce runtime policy before risky actions execute.
- Map blast radius across tools, data, and MCP servers.
- Send findings into Slack and SIEM workflows your SOC already runs.
- Generate executive- and compliance-ready evidence on an ongoing basis.
- Operationalize assurance with quarterly posture reviews and named support.
Enterprise
Starting at $500K per year
For regulated organizations that need private deployment options, advanced runtime protection, executive reporting, custom compliance evidence, and dedicated security support.
Outcomes
- Run in your cloud with BYOC, VPC, and dedicated data-plane options.
- Scale advanced Guardian workflows for regulated and high-assurance programs.
- Unify IAM with SAML and SCIM plus custom compliance mapping.
- Deliver board- and auditor-ready executive risk reporting.
- Cover 50–150 protected agents with contracted prompt volume.
- Align with a dedicated CSM, architecture review, and optional 24/7 SLA.
Building your first production AI agent? Ask about startup validation.
Enterprise program
Strategic AI Security Program
Starts at $1M plus per year
For Fortune 500, defense, banking, healthcare, government-adjacent, and critical infrastructure AI programs that need a dedicated operating model—not a standard product tier beside self-serve SKUs.
- Custom protected-agent and prompt volume
- Dedicated tenancy and deployment architecture
- Customer-controlled or dedicated data plane
- Dedicated support pod and executive sponsorship
- Custom detection packs and MCP protections
- Monthly or quarterly AI red team and board-ready reporting
- Custom procurement, legal, and residency programs
Questions before you buy
Short answers to the objections we hear during enterprise evaluations—before you scroll to the full FAQ.
What if we do not know how many agents we have?
Start with free discovery. You only pay when an agent is actively protected.
Need proof before buying?
Use a paid Production Validation POC. It credits toward your annual contract if you convert.
Need private deployment?
Enterprise supports BYOC and VPC options; Strategic adds dedicated tenancy and custom data planes.
Worried about prompt volume?
Prompt volume is included by tier and expandable by contract—we warn before sustained overages.
Professional risk reversal
POC conversion credit
Paid POC spend applies to your first annual agreement when you move forward.
Deployment success planning
Scale and Enterprise customers get a defined rollout plan with your CSM.
End-of-POC executive report
Paid and Enterprise POCs include leadership-ready reporting on findings and posture.
Security review support
Enterprise POCs include deeper support for architecture and control reviews.
Core protection included across Intertrace
Runtime Protection
- AI Security Gateway
- Real time prompt and response classification
- Prompt injection detection
- Jailbreak detection
- Tool call policy enforcement
- PII detection and redaction
- Secrets and credential detection
- Policy based allow, block, warn, redact, review
- Findings and evidence generation
Posture and Assurance
- AI asset discovery
- Protected agent inventory
- Per agent behavioral baselines
- Findings, incidents, and risk scoring
- OWASP LLM Top 10 mapping
- MITRE ATLAS mapping
- SOC 2 evidence exports
- Guardian monitoring
Autonomous AI Defense
Included capabilities by tier:
- LaunchGuardian Lite
- ScaleFull Guardian Agent
- EnterpriseAdvanced Guardian workflows
- Strategic ProgramCustom Guardian playbooks
Guardian continuously reviews runtime events, agent risk, tool calls, MCP activity, policy gaps, and evidence trails. It creates findings, recommends controls, and escalates issues based on customer approved guardrails.
Compare tiers
Feature depth is grouped so procurement and security partners can scan coverage, security capabilities, and enterprise readiness without one overloaded matrix.
Swipe to compare · feature column stays fixed
Coverage and scale
Agents, prompt volume, providers, clouds, and workspaces.
Security capabilities
Gateway, detections, MCP, Guardian, and blast-radius analytics.
Enterprise readiness
IAM, deployments, operations, and support.
Expansion options after initial deployment
Expansion is packaged so procurement teams see strategic coverage—not a scattershot menu. Use these paths when you need more agents, deeper integrations, dedicated infrastructure, or assurance programs.
MCP Security Pack
Scale integrations for tool- and MCP-heavy agents.
- MCP Gateway operationalization
- MCP Trust Center visibility
- Tool Guard and advanced approvals
- MCP risk reporting workflows
Compliance Evidence Pack
Map runtime evidence to the frameworks your GRC team cares about.
- SOC 2 AI control narratives
- NIST AI RMF mapping
- CMMC alignment where applicable
- EU AI Act support where applicable
- Executive-ready evidence exports
Private Deployment Pack
Bring Intertrace closer to your data and network perimeter.
- BYOC and VPC deployment options
- Dedicated or customer data planes
- Data residency programs
- Architecture review and success planning
Red Team Assurance Pack
Continuous adversarial testing tied to remediation and reporting.
- Quarterly or monthly AI red team cadence
- Scheduled autonomous probes
- Executive and board summaries
- Remediation workshops with your teams
Additional POC pathways
Production Validation is the fast path for teams that need a paid evaluation before annual contracts. These options extend scope for regulated environments or deep architecture reviews.
Standard POC
30 days included with annual contract
- Up to 3 protected agents
- Up to 1M prompts
- 1 provider
- 1 cloud
- Basic findings
- Evidence export
- Guardian Lite
- End of POC report
Paid POC
$25K to $50K
Credited toward annual contract if converted
Best for
- Security review
- Custom integration
- SIEM setup
- Procurement support
- Pre contract validation
Enterprise POC
$75K to $150K
Best for
- Regulated teams
- MCP Gateway validation
- Tool Guard validation
- SIEM integration
- Architecture review
- Executive report
- Security review support
Deployment options
| Option | Availability |
|---|---|
| SaaS control plane | Launch and above |
| BYOC | Enterprise |
| VPC | Enterprise |
| Dedicated tenancy | Strategic Program or Enterprise add on |
| Customer controlled data plane | Enterprise and Strategic Program |
Built for security review and procurement
- AWS Marketplace listing in progress
- SOC 2 evidence support
- SSO and SAML support
- SIEM integrations (Scale and above)
- BYOC and VPC options (Enterprise and Strategic)
- Architecture review availability
- Executive risk reporting (Enterprise and Strategic)
FAQ
Common questions from security, AI platform, and compliance buyers.
Secure your highest-risk AI agents first. Expand only when your AI footprint grows.
Book a pricing consultation to map protected agents, deployment options, and assurance cadence—or start with a paid POC that credits toward your first annual agreement.
Free discovery. Paid protection. Expansion that tracks real risk.
Free discovery. Pay only for active protection.